Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell: A Hurdle in Animal Life

Patrick H. O’Farrell

Department of Biochemistry, University of California San Francisco, San Francisco, California 94158 Correspondence: [email protected]

A requirement that an animal be able to feed to grow constrains how a cell can grow into an animal, and it forces an alternation between growth (increase in mass) and proliferation (increase in cell number). A growth-only phase that transforms a stem cell of ordinary proportions into a huge cell, the oocyte, requires dramatic adaptations to help a nucleus direct a 105-fold expansion of cytoplasmic volume. Proliferation without growth transforms the huge egg into an embryo while still accommodating an impotent nucleus overwhelmed by the voluminous cytoplasm. This growth program characterizes animals that deposit their eggs externally, but it is changed in mammals and in endoparasites. In these organisms, development in a nutritive environment releases the growth constraint, but growth of cells before gastrulation requires a new program to sustain pluripotency during this growth.

he phrase “Nothing in biology makes sense scribe four almost universal phases of growth Texcept in the light of evolution,” originally and proliferation in animal biology. I will be introduced in an essay supporting the theory focusing on the two earliest phases: the growth of evolution (Dobzhansky 1973), has been re- that produces a huge egg and the transformation purposed to chide biologists, who all too often of this single large cell into an embryo. The sub- ignore the origin and purpose of the biologi- sequent phases of growth were described in pre- cal processes and mechanisms they study. To vious reviews (O’Farrell 2004, 2011; O’Farrell look for how these processes “make sense,” one et al. 2004). But first, I will provide some context should look at theway theyserve the life histories for the comparisons that are made, and a brief of the organisms in which they are used. Such an synopsis of the four growth phases that convert examination can suggest how a process might an ordinary-sized stem cell into a big metazoan have been molded by evolution to achieve the adult. benefit provided. I have been an advocate for such considerations, because I believe them to MAMMALS CAME LATE be the best guide to interesting and important questions for investigation. Here, I will synthe- It is perhaps important to point out to the read- size observations made by examination of the er that I will first focus on animals that deposit natural histories of diverse organisms to de- their eggs externally, as this oviparous lifestyle

Editors: Rebecca Heald, Iswar K. Hariharan, and David B. Wake Additional Perspectives on Size Control in Biology: From Organelles to Organisms available at www.cshperspectives.org Copyright # 2015 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a019042 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042

1 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

is the most general and ancestral mode of ani- ALIGNING PROGRAMS OF DEVELOPMENT mal development. Mammals, of course, are in- AT A CONSERVED STAGE teresting to us, but if we wish to understand our connection to the rest of biology, we should The strategy used when comparing distantly re- first recognize the context within which mam- lated protein sequences suggests a general ap- mals appeared. Mammals are a small clade of proach when looking for distant homologies. 4000 species out of more than two million Rather than just aligning sequences at the amino animal species, and they are derived chordates terminus, one first identifies the most conserved that arose relatively late in animal evolution. domains and uses these for alignment. Similarly, Furthermore, mammalian evolution invested in comparing the programs of growth and pro- in a special program, which, by housing embry- liferation of diverse species, I have chosen to os in a nutritive environment, introduced an align the life histories of different organisms at unusual fashion of dealing with the growth con- a particularly conserved point (O’Farrell 2004; straints faced by all embryos. Understanding the O’Farrell et al. 2004). specializations of mammalian development in Theearlydevelopmental biologists,vonBaer this context can be illuminating (O’Farrell et al. and Haeckel, were fascinated by the similarities 2004). I will end with an effort to point out of vertebrate embryos of very diverse organisms these mammalian specializations and deep con- at a stage following establishment of the body nections to programs showed by their evolu- axis and formation of the neural tube. Recogni- tionary predecessors. But we will begin with a tion of this conserved stage and description of few guiding generalizations that give perspective the morphogenetic events that brought increas- on the coordination of growth with animal de- ingly dramatic distinctions to the embryos of velopment. different species were taken by some as evidence for a claim that ontogeny recapitulates phyloge- ny—or that the developmental sequence passes through all of the evolutionary steps as if “more AN ORGANISM MUST DEVELOP FEEDING advanced” organisms achieve their distinctions STRUCTURES BEFORE IT CAN INCREASE by late additions to the developmental sequence IN MASS (e.g., Graham andRichardson2012). Theattrac- Although there are some exceptionally large tive phrase, “ontogeny recapitulates phylogeny,” single-celled organisms, large body plans are appears to be somewhat of an overstatement and substantially the domain of multicellular organ- the idea is often described as discredited. None- isms. However, the production of a large body theless, the existence of a relatively similar em- from a single-cell zygote must deal with a fun- bryonic morphology is strongly supported. damental problem. Any animal whose nutrition A common embryonic morphology is also depends on a complex body plan with its spe- found among embryos within different inverte- cialized feeding structures must be able to de- brate phyla. This is particularly apparent in ar- velop these specialized structures before it can thropods. Arthropod embryos of very diverse feed and grow independently. In these cases, a species show remarkably similar morphologies complex body plan needs to be produced dur- on the initial establishment of the body plan. ing development before there is significant These similar looking embryos also use similar growth of the organism (O’Farrell 2004). molecules to guide the formation of embryonic We will view life histories of organisms in patterns (Patel et al. 1989; Prud’homme et al. the light of this problem, which appears to have 2003). This point in development has been acted as a constraint throughout the evolution called the phylotypic stage to reflect the fact of the animals. Despite great diversity, this van- that diverse organisms within a phylum show a tage point reveals relationships that lead to the “typical morphology” at this stage. global concept of four phases of growth and It is notable that the “phylotypic” stages of proliferation in the life plans of animal species. different phyla occur at a similar point during

2 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

embryogenesis, just after establishment of the same size, or relatively close (O’Farrell 2004). basic body plan, and embryos of even different Taking a little license with the exact stage that phyla have a very similar organization at this is defined as the phylotypic stage, the anterior/ stage. These distant similarities suggested by posterior axis of the phylotypic embryo is morphologies have been supported by molecu- roughly a millimeter. A few examples are a little lar analyses (Domazet-Losˇo and Tautz2010; Ka- smaller (e.g., Caenorhabditis elegans) or bigger linka et al. 2010). Key mechanisms patterning (e.g., grasshopper), and some embryos continue the phylotypicembryoareconserved across phy- to extend their length for some time after the la. For example, a gradient in a signaling mole- initial elaboration of the body axis. Nonetheless, cule of the transforming growth factor (TGF)-b sizes are rather similar. This similarity in size type guides dorsoventral polarity of the embryo, applies to the embryos of fruit flies, frogs, and and local expression of different homeotic-type blue whales, organisms whose adult masses, transcription factors subdivide the anteroposte- 1 mg, 100 g, and 100 metric tons, range over rior axis of the body plan (Shen 2007). These 1011-fold. This near-constancy in the size of findings suggest mechanistic parallels and argue the embryo at the phylotypic stage allows us to for evolution by descent. That is, even distantly divide our consideration of growth in animals related animals evolved from a common prede- into two. How do embryos of 10 mg(wet cessor with a similarembryonic stage, whichwas weight of cytoplasm) at the phylotypic stage patterned by mechanisms that continue to be grow to the adult sizes? How does oogenesis used in extant animals. Consequently, the phy- and early embryogenesis transform a single or- lotypic stage can be taken as an especially con- dinarily sized stem cell into a phylotypic em- served stage of development at which the embry- bryo, a miniorganism with a roughed-out body os of different species can be compared. plan? The original view exemplified by the “ontog- In this presentation, I consider growth as eny recapitulates phylogeny” phrase presumes an exponential process. Thus, the seemingly thatembryogenesisbeginswithacommonances- massive growth of a 7-ton blue whale calf into tral morphology that adds species-specific spe- a 150-ton adult is viewed as 20-fold and, cializations as embryogenesis progresses. Howev- hence, considerably less impressive than the er, the phylotypic stage occurs after important 100,000-fold growth of a normally sized early steps in embryogenesis and the modern stem cell to produce a 10-mg Drosophila egg. view considers development more like an hour- Examination of the life histories of numer- glass (Raff 1996; Prud’homme and Gompel ous species reveals four dramatically different 2010). Eggs are diverse, and distinctive programs phases of growth in the life plans of many ani- and morphologies are seen during very early em- mals: bryogenesis, but diversity shrinks with progres- 1. Sponsored growth: During oogenesis, a stem sion to the phylotypic stage and then reexpands cell grows enormously in size without divi- in later development. The bottleneck in morpho- sion. The mother sponsors this growth for logical diversity at the phylotypic stage presum- the benefit of the progeny. ably is a reflection of the early evolution of a suc- cessfulstrategyforbuildingabasicbodyplan,and 2. Subdivision: Following deposition of an egg the developmental reexpansion of diversity after in an external environment, the massive egg the phylotypic stage is a likely an indication that divides rapidly without any growth to pro- modificationandadditionofspecializationswere duce the phylotypic embryo. major routes of laterevolutionary diversification. 3. Expansion: Diverse but identifiable strate- gies are used in the growth of the phylotypic SUBDIVIDING ANIMAL GROWTH embryo to the adult. Embryos at the phylotypic stage are not only 4. Size control: Adult organisms usually con- similar in general morphology, but they are the strain their growth, a process that relies on

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 3 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

both quiescence and balanced replacement yolk vesicle of 17 g that is actually one extra- strategies. ordinarily large cell, the oocyte, before fertiliza- tion. Imagine a dinosaur oocyte! The chicken The central features of growth in phases 3 oocyte is more than a hundred-thousand- and 4 were described in two previous reviews million-fold (1011) larger than a typical cell, (O’Farrell 2004, 2011). Together with the con- which is 5 mm in diameter (or 10210 cm3, siderations presented here, these reviews reveal here approximated as 10210 g). On a geometric a perspective gained from examination of the scale, the 1011-fold growth to produce this large natural histories of organisms and their devel- cell in an ovary of the hen totally belittles the opment and growth. It is strikingly different roughly 100-fold growth of the hatchling to an from the views we develop when focusing on adult chicken. This growth of the single-celled some of our models of cell growth and prolifer- oocyte is, of course, growth without prolifera- ation. I suggest that the biological perspective tion. Although the growth of the yolk-rich avian gained in an exercise of considering how growth oocyte is above average, and so exaggerates the and proliferation are controlled in animal de- contribution of oogenesis to overall growth, it is velopment will uncover new and important fac- nonetheless true that, on a geometric scale, the ets of the control of these processes. growth of an oocyte makes huge contributions to the growth of organisms throughout animal PHASE 1—SPONSORED GROWTH: phylogeny. GROWING A HUGE CELL (OOGENESIS)

General Background Yolk versus Cytoplasm The development of an externally deposited egg Before embarking on a consideration of the into a feeding organism depends on supplies phases of growth, I would like to deconstruct received from the mother. This maternal dowry the extravagant process that occurs in the hen. has two parts: a large volume of cytoplasm, of- Besides familiarity, I use the chicken egg as an ten in the range of 100,000 times the amount example because of the dramatic size of the oo- found in a normal cell, and a supply of nutri- cyte. As mentioned, the size of the yolk supplies ents, largely in the form of yolk. The first growth provided by the mother differs widely between phase is maternally sponsored growth of a stem species, and it is the extravagant supply of yolk cell of ordinary proportions into a large oocyte that makes the avian oocyte a standout in size. that is often more than 105-fold bigger than an But chicken oocytes also have a very large cyto- average cell. This phase represents growth with- plasm, and here eggs of even very diverse organ- out division and it involves highly specialized isms are remarkably similar. In the case of the processes that are well studied in only a few chicken oocyte, the nonyolky cytoplasm forms systems. but a small whitish disc, the blastodisc, on the The vast majority of the millions of animal surface of the huge yolk. Although small on a species deposit their eggs in the external envi- relative scale, this small white area includes ronment in which, without further nutritional 10 mg of cytoplasm, a cytoplasmic volume input, they develop within protective shells or of 100,000 times that of an ordinary cell. The coating that largely isolate the embryo from its growth that produces this volume of cytoplasm environment. The duration and extent of the amounts to a remarkable achievement for one development that occurs during this autono- cell, and I consider this growth as the first of the mous stage varies substantially in proportion four phases of growth. to the nutritional supplies from the mother. After fertilization, this egg cytoplasm will From species to species, the amount of yolk divide rapidly to generate a cap of cells on top included in an egg differs over many orders of of the massive yolk vesicle, and these cells will magnitude. Avian eggs are at the upper end of undergo morphogenesis to produce the ele- this spectrum. For example, a chicken egg has a mentary body plan of the phylotypic embryo.

4 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

This subdivision of cytoplasm and production the rate of product synthesis—depends only on of a phylotypic embryo represents phase 2 of the the rate at which the polymerase traverses the growth and proliferation program, the subdivi- distance separating it from the previous poly- sion phase that is largely attributed to prolifer- merase. If we use 30 bp per sec as an approxi- ation without growth. For the moment, what I mation of transcriptional elongation rate and want to get across is that yolk contributes little 60 bp as the maximal packing density, a new to this process. To a large extent, it is the non- polymerase will advance to the end of the gene yolky cytoplasm that undergoes this phase 2 every 2 sec. Thus, at maximal loading, a single- program to produce an elemental phylotypic copy gene can produce a new transcript once embryo, and later the yolk is called on to fund every 2 sec. a third phase in the growth and proliferation A normally sized animal cell has 100,000 program, the expansion phase (O’Farrell 2004, messenger RNAs. The most abundantly ex- 2011; O’Farrell et al. 2004). Thus, we will ignore pressed genes in most cells represent several per- yolk in our considerations of phase 1 and phase cent of the total and would be present at levels 2 growths. We will instead focus on the cyto- between 1000 and 10,000 copies. At the max- plasmic accumulation in oogenesis and its sub- imal rate of gene expression, it would take be- division in early embryogenesis. tween 2000 and 20,000 sec, or between 30 min and 5 h, to make this amount of transcript, and half this time in a diploid and half again in a G Transcriptional Capacity, a Constraint on 2 cell, which has four copies of the genome. A Production of Large Cells growing cell must be able to produce a new com- There are numerous examples of huge cells: the plement of all of its mRNAs every time it dou- Drosophila salivary gland cells, the cells of Asca- bles. For many animal cells, this requirement is ris (a nematode that has a volume a billion-fold not a serious constraint because doubling times bigger than C. elegans but the same cellularanat- are usually considerably longer than the time omy), the giant cells of Aplasia, and ciliates of required to produce the most abundant tran- extraordinary size. But these cells have to either scripts. However, as a cell enlarges, the amount overcome or tolerate a biological constraint. The of mRNA in the cell increases in rough propor- genome has a limited ability to support growth tion to the cytoplasmic volume. Many oocytes or rapid changes in a large cell. As pointed out by achieve a size 100,000 times that of a normal Woodland (1982), the rate at which a gene can cell. It would take comparably longer to make produce transcripts is limited, and this limita- the transcripts to supply this large cytoplasm. tion imposes practical limitations on the growth Above, we calculated that a normal cell would and function of large cells. Let us look at this take 0.5 to 5 h to make its most abundant tran- limitation. scripts. If we increase the lower value 100,000- Rates of initiation of transcription cannot fold, we find that it would take 50,000 h be increased indefinitely. There is a packing lim- or 2000 d. For a G2 cell, with its four copies of it on the density of RNA polymerase molecules each gene, this reduces to 500 d. According to that can be loaded along a DNA molecule. Once this, the growth of cellsto avery large size should initiation rates are high enough to load a se- be an extremely slow process, and take roughly a quence to this limit, no further increase in the year. rate of RNA synthesis can be achieved. Irrespec- If an organism produces exceptionally large tive of the length of a gene, at the maximum cells, it must do so very slowly or find a solution packing density, polymerases are lined up one to the transcriptional rate constraint. In many behind the other at spacing estimated to be 50– cases, and in all of the examples of large cells 100 bp. A new transcript will be produced only given above other than oocytes, the genome is when a polymerase advances to the end of the amplified to increase coding potential. In ac- gene, and the rate at which each successive po- cord with the idea that transcriptional potential lymerase arrives at the end—corresponding to might limit growth, manipulations of the ploi-

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 5 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

dy of Drosophila salivary gland cells have shown plasmic constituents that are transferred to the that the size attained depends on ploidy (Fol- oocyte, which grows enormously over a few days lette et al. 1998; Hayashi and Yamaguchi 1999). at the expense of the nurse cells. But an increase in ploidy does not appear to be The nurse cell strategy is particularly effec- an option for the germline, as this would dis- tive in overcoming the limitation imposed by rupt the normal and faithful transmission of transcriptional capacity and it supports the rap- chromosomes from generation to generation. id increase in cytoplasmic volume of the devel- Ciliated protozoa have found a way to enjoy oping oocyte. The mother can also make other the benefits of polyploidy while retaining stable contributions to the development of the egg. genetics. Here, a disposable polyploid macro- Most notably, a different approach is generally nucleus supports large cells, while a distinct used to accelerate the accumulation of the yolky diploid germ nucleus, the micronucleus, bears nutrient supply of the oocyte. Other tissues of the responsibility for inheritance. How do oo- the mother, even distant tissues, such as liver, cytes in animals manage to grow despite the can contribute yolk protein, vitellogenin, lipids, limited transcriptional potency of their nuclei? and other nutrients that are taken up by the oo- cyte often with the assistance of surrounding follicle cells. The accumulation of yolk is of ma- Nurse Cell Support of Oocyte Growth jor importance to nutrition of the embryo and One effective way to speed the growth of a single later growth. The mother also provides further large cell is to enlist the help of other cells. A contributions in the elaboration of protective special organization is required if other cells are coatings and shells. But here we focus on the to contribute transcripts to a growing oocyte. In roughly 100,000-fold growth of the cytoplasm, many species of arthropods and annelids, the exclusive of yolk, as the germline stem cell ex- precursor of the oocyte divides, but only incom- pands to produce the oocyte. The numerous pletely, leaving sister cells interconnected by cy- copies of the genome available in nurse cells, toplasmic bridges that develop into specialized altogether a few thousand copies in the egg and stable intercellular conduits called ring ca- chamber of Drosophila melanogaster, provide nals. The interconnected group of sister cells is abundant template for rapid accumulation of called a cyst when it first forms, and, in insects in transcripts and growth of the oocyte. In organ- which it is a stable unit in oogenesis, it is called isms that use this strategy, a stem cell can mature an egg chamber once it is invested in a coating of into a large oocyte relatively quickly, 10 d in independently derived follicle cells. Although the case of Drosophila. one cell of the interconnected group becomes the oocyte, the remaining interconnected cells Autonomous Growth of Oocytes serve as nurse cells that supply the oocyte. The nurse cells amplify their genomes and their ca- Cytology and electron microscopy have been pacity to produce RNA in endoreduplication cy- used to survey oogenesis in numerous organ- clesconsisting ofdistinctrounds ofSphasewith- isms. These analyses define different oogenesis out cell division. The resulting large polyploid strategies. In many species, the precursors of oo- cells efficiently fuel the growth of an immense cytes either lack or lose connections with sister eggcytoplasm.Forexample,inDrosophila,apre- cells and the individual oocytes grow autono- cursorcell called acystocytedividesfour timesto mously, often with an isolating coating of folli- create an interconnected “cyst” of 16 cells that cle cells that do not have detectable cytoplasmic develops into an egg chamber in which 15 of continuity with the growing oocyte. In these or- the cells become highly polyploidy, and one ganisms, the oocyte nucleus appears to support cell, which will grow into the oocyte, arrests in the growth in cytoplasmic volume, and the con- meiotic prophase. Although the oocyte nucleus tributions of the maternal environment appear remains transcriptionally inactive, the nurse to be limited to provision of yolk and other nu- cells produce an abundance of RNA and cyto- trients and production of protective coatings.

6 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

Given the limitation imposed by the tran- Selective Amplification of Genes Encoding scriptional potency of a single nucleus, how Ribosomal RNA does autonomous growth succeed at producing oocytes? To a large degree, the answer is that Unlike protein-coding RNAs whose impact oocytes using this program grow much more is further amplified by translation, structural slowly, generally taking several months to more RNAs, such as ribosomal RNA, function di- than a year to reach a mature size (Smiley 1990). rectly, and there is no way to bypass or minimize But even this pace appears to rely on several the need for a high transcriptional capacity to striking specializations that appear designed to accumulate a high level of functional product. enhance provision of key transcripts. Furthermore, there are a lot of ribosomes in even As expected, when growth is autonomous, an ordinary cell, roughly several million (Blobel the oocyte nucleus is transcriptionally active and Potter 1967; Wolf and Schlessinger 1977), a (Gall et al. 2004). The growth phase still occurs few orders of magnitude more than most abun- during an arrested prophase of meiosis I, but dant mRNAs. If there were only one gene encod- meiosis is modified to incorporate a special ing ribosomal RNA, the nucleus would take transcriptional phase generally not seen in sys- about a month to meet the transcriptional de- tems using nurse-cell-supported growth. Meio- mand required to reproduce a normally sized sis begins normally. The chromatids condense cell. However, ribosomal RNA is encoded in re- on the start of prophase, and progress through peating arrays of genes at loci often called nucle- the meiotic events of homolog pairing and re- olar organizers. The repeats increase the tran- combination before a diplotene arrest. At this scriptional capacity, but phenotypes resulting arrest, large loops of chromatin unfurl from a from partial deletions of the recombinant DNA still-visible chromosome axis. The axes outline (rDNA) repeats in Drosophila suggest that, even the meiotic arrangement of paired chromo- with this repetition, there is little reserve capac- somes, suggesting a change in chromatin com- ity (Gersh 1968). In species in which oocytes paction without disruption of the meiotic chro- grow autonomously, the coding capacity of mosome configuration. The chromatin loops the rDNA repeats would be severely taxed and are packed with RNApolymerase and elongating should limit growth. However, a remarkable transcripts giving a striking appearance noted by process takes care of this limitation. early microscopists. Because they resembled the As oocytes progress through meiotic pro- long fuzzy brushes that were once used to clean phase to their arrest in diplotene, they hugely the soot from the glass chimneys on oil-burning amplify the ribosomal repeats (Brown and Da- lamps, they are called lampbrush chromosomes. wid 1968; Gall 1968). The replication associated This stage often persists for months asthe oocyte with this amplification has been detected by enlarges, and studies in Xenopus showed that labeling, and the DNA copy increase was as- oocyte transcripts accumulate during this peri- sessed by molecular and cytological methods. od (Gall 2012; Kloc et al. 2014). This amplification can be so extensive that it The lampbrush chromosome stage provides increases the total amount of DNA in the nu- a long period of highly active transcription with cleus several-fold. The amplified sequences are a postreplicative nucleus that has four copies of produced as free circles that form hundreds of the genome. Still, our rough calculations would small nucleoli throughout the oocyte nucleus. suggest that a very long time would be required Amplification occurs in diverse organisms that for the transcript accumulation needed for a use oocyte-autonomous growth and is pre- 100,000-fold increase in cytoplasmic volume. sumed to support the massive increase in ribo- It would not be surprising if there were addi- somes during growth. tional adaptations that might reduce the severity We have seen three striking biological of the transcriptional constraint so that oocytes innovations that promote growth of oocytes: could be produced more rapidly. Indeed, there is nurturing by nurse cells, transcription from dramatic evidence for such adaptations. lampbrush chromosomes, and amplification

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 7 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

of rDNA. Apparently, the nurse-cell strategy re- of the germinal vesicle, and forms an easily duces the need for other adaptations to increase stained compact structure called the karyosome transcriptional output, as oocytes supported by within a sea of nucleoplasm. Avariety of special- nurse cells appear to be transcriptionally silent ized structures have been described in the ger- and to lack lampbrush chromosomes (Ganot minal vesicle (Gall et al. 2004; Gardner et al. et al. 2008). These relatively widespread special- 2012). One of these structures, the mininucleoli, izations modify three fundamental biologi- which appear throughout the germinal vesicle, cal processes. First, cytokinesis is rendered in- results from the amplification of ribosomal re- complete and the arrested cleavage furrows are peats and is clearly a response to the dispropor- transformed into ring canals that interconnect tionate demand for ribosomal RNA synthesis nurse cells and oocytes. Second, although tran- described above. scription is normally discontinued on mitotic The growth/multiplication of mitochon- or meiotic chromosomes, the lampbrush chro- dria in the oocyte faces huge distortions in the mosome stage is created by reactivating tran- balance of different mitochondria gene prod- scription of condensed meiotic chromosomes. ucts, those encoded in the nucleus and those Finally, although DNA replication is ordinarily encoded by the mitochondrial DNA. The mito- tightly regulated to ensure that all sequences are chondrial genome is replicated in proportion replicated once and only once, these controls to the expansion of the cytoplasm. In Xenopus, are bypassed to massively amplify the rDNA 7 108 copies of mitochondrial DNA are pro- repeats. These are not oddities of a peculiar duced during oogenesis (Chase and Dawid branch of evolution—they are modifications 1972), whereas the nuclear DNA is unchanged. making key contributions to a central step in This represents nearly a million-fold increase in the life history of nearly all animal species, the mitochondrial genome copy number above that production of an oocyte. in a typical cell. Because nuclear and mitochon- drial DNA–encoded gene products assemble stoichiometically to make the respiratory com- Unbalanced Growth plexes, continued production of functional mi- The accepted truism about cell growth is that all tochondria must be managed in the face of a of the contents of a cell need to double, at least changing imbalance in the two contributions. on average, every time a cell divides. This applies We know nothing of the mechanisms that deal to the much-studied case of exponentially grow- with this supply imbalance, one that is faced to a ing cells in culture. However, growth to produce smaller degree in other enlarging cells, such as an organism is not simply a matter of increasing neurons. everything equally. Growth in vivo is often un- The great bulk of the multiplication of mi- balanced, and, at a cellular level, this is particu- tochondria occurs in a specialized structure, the larly dramatic when one considers the huge Balbiani body. This structure, which is found in growth of a stem cell into an oocyte. Extraordi- the oocytes of a wide range of animals, is seen as nary changes occur in the size of some struc- a differentiated area of cytoplasm that can be tures, and the stoichiometry of some key cellular detected early in oogenesis, just as precursor components. As I describe here, this places huge cells are committing to oogenesis (Kloc et al. stress on growth itself as a result of dispropor- 2014). It is usually adjacent to the nucleus. Of- tionate changes in pathways contributing to ten, it is a dominating structure obvious in the growth. light microscope. In the electron microscope, The nucleus of the oocyte is huge—so big it is seen as a congregation of mitochondria, that it has its own special name, the germinal endoplasmic reticulum, and an RNA-rich struc- vesicle. Despite having only the DNA content of ture called nuage (French for “cloud”). Nuage a normal G2 cell, the nucleus expands in rough appears to be universally associated with de- proportion to the entire oocyte, 100,000 veloping oocytes and more generally with times. The DNA occupies only a tiny volume the germline. Localization, genetic dissection

8 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

and identification of molecular constituents the 100,000-fold increase in the number of mi- have tied nuage not to one, but to several func- tochondrial genomes occurs in a few days in this tions—storing mRNAs for the future embryo, more rapid style of oogenesis. specifying the vegetal or posterior pole of One final feature of mitochondrial multipli- the egg, defining the germline, and protecting cation dramatically emphasizes the fact that against infectious DNA elements by housing an growth in vivo is “unbalanced”—that not all immunity system based on a category of small cellular components double when cells double. RNAs, piRNAs (Marlow and Mullins 2008; Kloc In C. elegans, a supply of 100,000 mitochon- et al. 2014). Obviously,therearemany tasksto be drial genomes is accumulated during oogenesis, performed in the oocyte, and rapid progress is and no further increase occurs until nearly adult being made in defining the contributions of stages. Additionally, a mutant lacking the late nuage. But here, we focus on the mitochondria. increase in mitochondrial genomes produces The cytoplasmic Balbiani body of Xenopus, mature adults (Tsang and Lemire 2002; Bratic which is densely packed with mitochondria, et al. 2009). Thus, the supply of genomes made grows to be huge. Its increase in size is paralleled during oogenesis suffices for the life of the by an increase in mitochondria and mitochon- worm. It is likely that the behavior seen in C. drial DNA, and the structure only disperses elegans is broadly representative, although likely months later when the oocyte turns its attention less extreme in other systems. After the huge to yolk accumulation. Mitochondrial morphol- increases in mitochondrial DNA copy number ogy and DNA labeling argue that Balbiani body during oogenesis in Xenopus and in Drosophila, mitochondria proliferate (Chase and Dawid little or no further increase occurs during all of 1972; Webb and Smith 1977; Kloc et al. 2014). embryogenesis (Chase and Dawid 1972; H Ma Indeed, it appears that more than 16 doublings and PH O’Farrell, unpubl.). of the mitochondrial genome occur before dis- Although these descriptions of mitochon- persal of the Balbiani body in Xenopus. drial multiplication during oogenesis do not re- The Balbiani body is not always a dominat- veal mechanism,they give averydifferent viewof ing morphological feature. In Drosophila, it is what mechanisms we should be looking for. Mi- difficult to see by light microscopy without spe- tochondrial genome replication is uncoupled cialized tags, and while obvious by electron mi- from cellular growth and proliferation, but is croscopy, it is rather transient in comparison to strikingly regulated during development and its persistence in autonomously growing oo- shows a remarkable departure from balanced cytes (Cox and Spradling 2003). This is likely growth. To understand how the copy number linked to delegation of some responsibility for of this genome is regulated, we need to look for production of various components to the mul- developmental inputs, and we might also expect tiple nurse cells. In earwigs, an insect wherein a special regulatory circuits that handle the rapid single large nurse cell supports each oocyte, a multiplication in the face of a dramatically Balbiani body is found in the nurse cell as well as changing context during oocyte growth. the oocyte (Tworzydlo et al. 2009). Nurse-cell support of oogenesis, as exempli- Constrained Change versus Conservation fied by Drosophila, reduces one type of stress. The increased nuclear transcription capacity The specialized features described above can be provided by the polytene nurse cells reduces found in diverse organisms with representatives the imbalance in nuclear and mitochondrial in more than one phylum. For example, lamp- copy number. However, the transcription of nu- brush chromosomes have been described in or- clear products occurs in a distinct, albeit inter- ganisms representing mollusks, arthropods, connected, nurse cell compartment. Now, there echinoderms, and chordates (Bedford 1966; is a need to manage the flow of gene product to Gruzova and Batalova 1979; Smiley 1990; Gall the oocyte where mitochondria are actively rep- et al. 2004). This gives the impression of conser- licating. Perhaps more impressive is that most of vation, and conservation often comes with a

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 9 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

connotation of stability. However, change is per- scripts for growth of the oocyte, but eventually sistent, and it imposes a constant diversifying all but one germ nucleus is eliminated (Del pressure that is resisted by selection. Constraints Pino and Humphries 1978; Macgregor and Del imposed by a requirement for function limit Pino 1982). Also, nematodes produce eggs from some change, but not all. Thus, like the cars a syncytial gonad in which many nuclei contrib- produced over the past several decades, which ute to a cytoplasm that is partitioned into the continue to have four wheels (generally), a rapidly formed large egg (O’Farrell 2004). Thus, steering mechanism and, brakes, a lot of changes helpercells,orhelpernuclei,promotethegrowth occur “under the hood.” Variations in the for- of large oocytes by means of shared cytoplasm mat of oogenesis and the distributions of the in at least a few animal phyla. This might repre- variants in phylogeny reveal changes, but the sent convergent evolution to this strategy. retained features suggest operation of powerful An alternative to convergent evolution is and lasting constraints that limited the range of divergence from an unrecognized common pro- successful formats. genitor. Indeed, at least one observation might Although our sampling of species is still rel- suggest a deep connection between nurse-cell- atively small, the phylogenetic distribution of supported oogenesis and autonomous oogene- nurse-cell-supported oogenesis versus autono- sis. In numerous organisms, cytokinesis of the mous oogenesis is of interest. Nurse-cell-sup- precursors to the oocyte is incomplete and pro- ported oogenesis is seen in numerous insect spe- duces specialized bridges interconnecting the cies and has been particularly well studied in sister cells (Pepling et al. 1999; Marlow and Mul- Drosophila; however, there are also many insects lins 2008). In organisms with nurse-cell pro- including members of one of the most diverse grams, only one cell of the interconnected group groups of animals on earth, the beetles, which becomes the oocyte, and the bridges persist so show autonomous growth of their oocytes. Be- that the other cells, the nurse cells, can transmit cause the nurse-cell format of oogenesis is their cytoplasmic endowment to the oocyte. found, at least so far, in derived insect lineages, However, in organisms showing autonomous it is suspected that this program itself is derived. oocyte growth, the relationship is transient, is This interpretation is consistent with findings of lost before oocyte growth, and usually each of autonomous oocyte growth in sampled species the originally connected cells makes an oocyte. of various other animal phyla, including chor- Although a common progenitor to nurse-cell dates, echinoderms, mollusks, and cnidarians, programs and autonomous oogenesis has not and a group of more primitive animals, such as been recognized, we should not forget that sex jellyfish (Smiley 1990; Eckelbarger and Larson appeared earlier in evolution than animals. Per- 1992; Eckelbarger and Young 1997; Marlow and haps vestiges of early gametogenesis programs Mullins 2008). However, the nurse-cell program provided a foundation for the different strate- is not limited to derived insect lineages. Nurse gies present in animal species today. cells are also used in annelids, and a related Regardless of the path taken in evolution, nurse-cell type of oogenesis has been discovered representative animals in every major branch in a group of chordate species whose lifestyle of phylogeny grow large oocytes, and these var- relies on rapid oogenesis (Ganot et al. 2007). ied organisms call on a limited set of specialized Furthermore, a group of frogs that carry their mechanisms to support this growth. I suggest eggs in a brooding pouch, marsupial frogs, that this reflects the persistent operation of con- does oogenesis differently than other frogs, and straints that guide selection. An operational the part that is especially different has features necessity to produce an egg that can support like nurse-cell-supported oocyte growth. In development to an autonomous feeding stage these frogs, oogonia undergo multiple incom- might have been the driver of other constraints plete divisions to create a dramatically multinu- as follows. Development of a feeding organism cleate oocyte: these nuclei develop lampbrush without growth demanded a large egg. Produc- chromosomes, apparently all contributing tran- tion of a large egg required extensive growth of

10 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

the oocyte. Production of very large cytoplasm 2. Slow accumulation of RNA was a chief limi- harboring only a single nucleus distorted the tation to the growth of a huge oocyte, and now usual biological relationships, and the resulting the egg has this same cumbersomely large cyto- stresses produced selection for special adapta- plasm and a single nucleus. Slow accumulation tions. of transcripts would stymie the progress of de- velopment (Woodland 1982). As a result, eggs initially avoid relying on transcription. PHASE 2—FROM EGG TO EMBRYO The egg immediately embarks on a program Fertilization and egg deposition marks a major that will resolve the discordance between the turnaround for growth. After growing to huge cytoplasmic volume and the nucleus. Rapid dimensions as a single cell, nutrients are no lon- cell cycles exponentially increase the number ger freely available as the egg sets out to make an of nuclei, whereas the cytoplasmic volume re- embryo. In doing so, the huge egg is subdivided mains constant and so reduces the cytoplasm into many cells and is restructured, initially for which each nucleus is responsible. As the without growth. It is not a time to tarry. The nuclear to cytoplasmic ratio increases, tran- egg is a defenseless immobile package of nutri- scription can, and does, begin to play an impor- ents, and speed of development is one way that tant role in regulation. externally deposited eggs minimize predation. Importantly, the early cell cycles that ramp We tend to think of development as a rela- up the transcriptional potential of the embryo tively protracted affair, perhaps because of our are themselves independent of transcription. own gestation time. But the long periods of fetal Regulated gene expression normally plays a ma- growth in mammals are more about growth than jor role in the control of a whole host of biolog- development, as is the time of yolk-supported ical processes, including regulation of the cell growth of birds. The relationship between devel- cycle, and, by forsaking transcriptional control, opmental time and growth is perhaps most ob- the embryo has to extensively restructure bio- vious in birds, in which the time to hatching in logical regulatory mechanisms. Consequently, different species increases in proportion to the the egg, at the starting block of development, size of the egg and the hatchling it eventually uses unusual biological mechanisms. As I will produces (Rahn et al. 1974). Externally depos- discuss below, the unusual aspects of this ances- ited eggs with little yolk often develop quickly. tral program of fast, transcription-independent C. elegans develops from a fertilized egg into a cell cycles has a pervasive impact on the biology functional organism in 13 h. Some clam species of the early embryo and influences the ground- hatch as ciliated trochophore larvae within 24 h, work on which subsequent development has a Drosophila larva crawls away from its eggshell been built. But first, I will very briefly review 24 h after fertilization, and zebrafish swim after more studied aspects of this early embryonic 3 d. Apparently, the extraordinary processes program. of morphogenesis and differentiation that build an organism do not need to take a long time, Rapid Cell Cycles and the Midblastula but growth does. To sidestep variations associ- Transition (MBT) ated with growth, we will consider development from fertilization to the time of the phylotypic The rapid early embryonic divisions have been embryo as the second phase of “growth” and an attractive model for cell-cycle studies, and proliferation. I have called this phase “subdivi- the return to reliance on transcription has sion,” because it primarily involves subdivision been an interesting area of developmental anal- of the mass of the oocyte, without growth (i.e., ysis. Fast, early cell cycles are widespread. without a substantial increase in mass). The mitotic cycles of newly fertilized eggs of As we saw for phase 1 growth, the limited Drosophila, sea urchin, and Xenopus are 8.6, transcriptional potency of a nucleus also con- 30, and 30 min, respectively. These cycles lack strains early developmental events during phase gap phases, and occur even if transcription is

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 11 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

prevented—indeed, in frog and sea urchin, they relatively short time span that is called the mid- have been shown to occur even if the nucleus is blastula transition (MBT) (Farrell and O’Farrell eliminated (Harvey 1935). After several period- 2014). ic divisions, the cycles begin to slow, at first Although it is not yet understood how the gradually and then more abruptly, as the em- MBT is controlled, it seems natural to believe bryo approaches gastrulation. that the embryo is running out of something— In addition to its transcriptional indepen- perhaps a key component is used up, or a ma- dence, the early cell cycle is redesigned in several ternal supply becomes inadequate as nuclei in- ways to achieve speed. It lacks the gap phases, G1 crease. Experimental manipulation of the ratio and G2, which usually separate S phase and mi- of nuclei to cytoplasm suggests that its increase tosis. Shortening of S phase by as much as 100- has an input into the MBT (Newport and fold further quickens these cycles (Farrell and Kirschner 1982a,b; Edgar and Schubiger 1986; O’Farrell 2014). These features of the early cell Edgar et al. 1986; Pritchard and Schubiger 1996). cycle are seen in diverse settings—the rapidly Progress has been made in understanding how dividing teleoblasts of annelid embryos, the the cell cycle slows. For example, inhibitory syncytial mitotic cycles of Drosophila, the early phosphorylation leads to a decline in activity divisions of sea urchin, and those of Xenopus. of the cell-cycle kinase, Cdk1, prolongs S phase, Thus, these cycles, which are often viewed as pe- and introduces a G2 phase at the MBT in Dro- culiar and distinct from “normal,” are used dur- sophila (Edgar and O’Farrell 1990; Farrell and ing the initial steps of development of diverse O’Farrell 2013). Increasing nuclear density ap- animals. I suggest that the constancy of these pears to one of the signals triggering slowing of features is the result of constraints demanding the cell cycles; however, the mechanism of this rapid expansion of transcriptional potential in coupling remains unclear. It is also uncertain the unwieldy large-egg cytoplasm. whether the nuclear to cytoplasmic ratio acts One of the most fascinating questions re- directly on other MBT events, or whether its garding these early cycles is: What terminates effects are mediated by changes in the cell cycle. them? The egg is primed by its maternal dowry to initiate relentless and rapid cycles that have Distinctive Regulation during Early Cell Cycles no obvious outside input. But the rapid cycles have to end, and they slow in a stereotyped fash- Beyond suggesting that growth issues underlie a ion just before gastrulation. Furthermore, the universal program of early cell cycles without maternally deposited gene products need to re- growth or transcription, creation of this special linquish control to transcription of the zygotic proliferation stage has been associated with a genome. This changeover, which involves de- major retooling of regulation so that embryonic struction of maternally provided RNAs and development begins in a special context. proteins and activation of zygotic transcription, In addition to impressive diversity in na- is referred to as the maternal-to-zygotic transi- ture, evolution creates distinctions between tis- tion (MZT). In part, this is a progressive tran- sues and between stages in the life histories of an sition with additional zygotic functions coming organism. Exemplifying the latter, evolution into play as persisting maternal functions are optimizes larval and adult forms for different successively eliminated (Wieschaus 1996; Fol- lifestyles. Phase 2 of animal growth is present in lette and O’Farrell 1997). Nonetheless, during the life histories of all animals, and evolution this process, there are abrupt transitions when might have optimized this stage. As evidence for the nearly silent nuclei activate expression of such an evolutionary trajectory, I present three new panels of genes (Newport and Kirschner aspects of regulatory biology that are widely 1982a,b; Edgar and Schubiger 1986; Edgar et conserved among eukaryotes, and yet are not al. 1986). The slowing of the cell cycle, MZT, followed by early embryos. and onset of the cell-shape changes and move- Cells coordinate the synthesis of the precur- ments that initiate morphogenesis occur over a sors of DNA, the deoxynucleotide triphosphates

12 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

(dNTPs), with replication. By providing the absent, and all sequences replicate at the same dNTPs just as they are used, their concentration time (Shermoen et al. 2010; Li et al. 2014; Yuan never rises very high, and hydroxyurea inhibi- et al. 2014). Thus, the specialization of chroma- tion of ribonucleotide reductase, an enzyme key tin into euchromatic and heterochromatic do- to producing dNTPs, quickly brings DNA rep- mains, a feature of regulatory biology that spans lication to a standstill in organisms from bacte- known eukaryotic biology, is abandoned in the ria to human. Things are different in the early early embryonic cell cycles. embryo. Xenopus eggs have high levels of dNTPs It is dangerous to suggest why things are the (Woodland and Pestell 1972). Additionally, way they are, but it seems likely that the dramatic hydroxyurea does not block the early S phases distinctions in mechanisms that characterize of Xenopus or of Drosophila (Landstro¨m et al. the early embryo are secondary to the devoted 1975; Newport and Dasso 1989; Howe et al. attention to the pell-mell cell cycles. Cell-cycle 1995; K Yuan, AWShermoen, PH O’Farrell, un- events can disrupt other processes. The spindle publ.). Sea urchin eggs initiate cleavage cycles usurps cytoskeletal components, and cytokine- without one of the subunits of ribonucleotide sis disrupts intercellular membranes. Further- reductase and must similarly have a preformed more, nascent transcripts are aborted during pool of dNTPs (Evans et al. 1983). Thus, eggs in mitosis, interrupting ongoing transcription un- at least three different phyla deviate from an til new RNA polymerases transit the entire gene otherwise universal program of regulation and (Shermoen and O’Farrell 1991). In the early carry a pool of dNTPs that provides DNA pre- embryonic cycles, cells are in mitosis for a large cursors for all of the early cell cycles. fraction of the cycle, and many cell biological Histones assemble newly replicated DNA processes appear to be held in abeyance. into chromatin, and, like the dNTPs, histones In summary, constraints forced an alterna- are usually made only during S phase. Again ear- tion between different styles of growth. Each ly embryos are different. The egg carries a stock- style of growth elicited evolutionary adapta- pile of histone proteins. The egg also has a large tions that altered fundamental features of bio- capacity to store this depot in complex with logical regulation. Consequently, the transitions chaperones, such as nucleoplasmin (Laskey et from one growth phase to another are widely al. 1978). This deviation from an otherwise uni- transformative. versal program regulating histone supply could have a major consequence on the chromatin as- Erasing the Slate for Development sembled. Two types of chromatin exist in eukaryotes Becauseanimalsevolvedwithalternatinggrowth from yeast to humans: heterochromatin and phases as a consistent context, developmental euchromatin. The especially compacted het- processes might be optimized to function in erochromatin, originally defined by intense his- their phase-specific biological setting. In this tologic staining, is now more commonly dis- way, the unusual biology of the proliferation- tinguished by molecular hallmarks (e.g., a only phase might have become an integral part histone modification, H3-K9 methylation, and of programs governing early animal develop- binding of heterochromatin protein 1 [HP1]). ment. Here I consider pluripotency—the ability These marks are considered repressive because of cells to form multiple cell and tissue types. heterochromatin is generally transcriptionally Induction of pluripotent stem cells in mam- quiescent. Additionally, heterochromatic DNA mals is highly topical as result of current ex- replicates later than euchromatic sequences, citement about stem cells and their therapeutic and the sequential replication of different re- potential. However, the capacity of a cell to pro- gions of the genome is the main reasons that S duce diverse cell types is broadly important. As phase is normally long (Shermoen et al. 2010). development progresses and cells differentiate, But during the early cell cycles, chromatin is their genomes and the chromatin that packages relatively decompacted, repressive marks are these genomes are modified to restrict gene

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 13 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

expression. But all life forms must begin each sperm nucleus, which then failed to contribute generation anew with a totipotent progenitor. to zygotic events (Loppin et al. 2005). Hira is a Though the epigenetic marks that specify conserved member of a protein complex that the fates of cells can be transmitted over many acts as an H3/H4 chaperone to promote repli- cell divisions, John Gurdon’s classical experi- cation-independent deposition of nucleosomes ments showed that a nucleus of a differentiated (Orsi et al. 2013). The Hira homologs in fish cell, when transplanted into an enucleated frog and mouse are similarly required for decompac- egg, could support development of a viable and tion (Zhao et al. 2011; Lin et al. 2014). fertile frog (Gurdon et al. 1958). Thus, the egg, Decompaction is a general activity of egg acting either directly or by driving the implant- extracts with extracts of one species being able ed nucleus through a rapid series of maternally to decompact the sperm of various species, de- programmed divisions, can expand the devel- spite differences in the SNBPs. Furthermore, opmental potency of an otherwise inadequate these extracts will decompact the chromatin of nucleus. other nuclei, such as the inactive nuclei of avian The resetting of epigenetic marks is a com- red blood cells. I suggest that the ability to ex- plex topic as there are many types of marks and change basic chromatin proteins, whether prot- complex interactions influencing their produc- amines or histones, with a pool of naı¨ve histones tion, inheritance, and reversal. It is now known contributes to the ability of the egg to repro- that pluripotency can be established by experi- gram the epigenetic state of a nucleus (see Lin mental treatments, but the egg cytoplasm is spe- et al. 2014). cial in that it is both capable of resetting epige- In contrast to egg extracts, oocyteextracts are netic marks and is the natural environment in ineffective at sperm decompaction. Nucleoplas- which such a resetting should occur. To get a min is extensively phosphorylated at the time of view of how the egg might reprogram a nucleus, egg activation, and dephosphorylation of nucle- I consider its specialized ability to activate one oplasmincompromisesitsactivity indecompac- of the most repressed and condensed nuclei tion (Philpott et al. 2000). Phosphorylation of known, that of the sperm. nucleoplasmin persists until the MBT is depen- Following fertilization, the highly con- dent on high-cyclin-dependent kinase activity, a densed sperm nucleus swells and the chromatin specialization of the proliferation-only phase. decompacts, replicates, and forms the male pro- Additionally, the egg and early cleavage stage nucleus, which joins the female pronucleus to embryo are well positioned to reprogram epige- participate in the rapid early-division cycles. In netic marks because they have especially high the early stages of this process, sperm nuclear levels of the histone chaperones and a unique basic proteins (SNBPs), which include special- pool of naı¨ve histones available for exchange. ized histones and protamines, are replaced with The speed of the early embryonic cycles more conventional histones. Xenopus egg ex- might also contribute to reprogramming by tracts promote this decompaction of sperm nu- outpacing the mechanisms responsible for epi- clei. Immunodepletion showed that this activity genetic inheritance. Histone methylation and depends on nucleoplasmin, the abundant his- DNA methylation marks are added after DNA tone chaperone (Philpott et al. 1991). Nucleo- replication and histone deposition, and, if cells plasmin-like chaperones in flies, sea urchins, progress immediately to mitosis and another and frogs appear to be involved in sperm de- round of replication, there may not be adequate compaction. time to modify the newly replicated DNA and The octomeric nucleosome is built in a two- newly deposited histones. Indeed, during the step process by deposition of H3/H4 tetramer extremely rapid early cell cycle in Drosophila followed by addition of a H2A/H2B tetramer. embryos, chromatin lacks hallmarks of hetero- Nucleoplasmin interacts with H2A/H2B, and chromatin, H3K9 methylation, and bound HP1 so appears specialized for the second step. A binding, and these only make their appearance fly mutant, hira, was unable to decompact the as the cycles slow. When the features of hetero-

14 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

chromatin do appear, they do so in a sequence of alignment to earlier stages reveals a point of as if heterochromatin formation is an integral divergence in the programs. In all three organ- part of the developmental program (Shermoen isms, gastrulation precedes establishment of the et al. 2010). phylotypic body plan and is regulated by related I suggest that erasure of the epigenetic marks molecules and mechanisms, even if differences in externally deposited eggs is coupled to the in topology of the movements obscure the par- unusual context in the egg and early embryo, allel. In frog and chicken, as well as avast number and that restoration of epigenetic marks awaits of organisms throughout phylogeny, the rapid slowing of the cell cycle when chromatin be- cleavage cycles of the egg immediately precede comes less dynamic and can again accumulate gastrulation so that the blastomeres produced by modifications. As discussed below, even though cleavage make up the gastrula. In contrast, cleav- these mechanisms are set in a different context age of the mouse egg begins 6.5 d before the start in mammalian development, an understanding of gastrulation, and the inner cell mass (ICM) of the ancestral mechanism will give us deeper cells and trophoblast cells that result from these insight into this process in mammals (Can˜on cleavages do not initiate gastrulation (Fig. 1), et al. 2011; Sa´nchez-Sa´nchez et al. 2011; Ma- but instead begin a growth program discussed randel et al. 2012; Lee et al. 2013). below. Recognition of the changes associated with the insertion of the mammalian specific module gives us new ways of thinking about WHY ARE MAMMALS SO DIFFERENT? the parallels between organisms. Mammalian eggs are smaller than externally de- posited eggs, and their early cleavages are slow. Insertion of Mass Increase into Phase 2 So, despite a long heritage of predecessors that Growth adhered to the four phases of growth that I have described, the early growth phases of mamma- Protected and fed by the maternal environ- lian embryos are altered. Here, I explore the ment, the mammalian embryo is released nature of the difference, and why the mamma- from growth constraint, and its altered devel- lian program differs. opment depends on the relaxation of the con- straint. Mammalian eggs are small, and have 100–1000 times less cytoplasm than a typical The Alignment Problem externally deposited egg. Consequently, mam- One of the most dramatic oddities of mamma- malian embryos need to grow to produce a lian development is that it does not start at the phylotypic embryo of ordinary proportions. A same point asthe development of other animals. growth period added to embryogenesis pro- A developmental module generating extra em- vides the chance to “catch up” to the embryos bryonic tissues is inserted at the beginning of of externally deposited eggs. mammalian development. This insertion ob- The early mammalian egg begins develop- scures relationships even with close vertebrate ment at a leisurely pace. One to twocell divisions relatives (O’Farrell et al. 2004). Certainly, when per dayare typical in early mammalian embryos. I was faced with descriptions that began with This slow beginning is a feature that appears in fertilization, I was hard pressed to see common the derived mammalian lineage and is not seen features among of the developmental programs in closely related vertebrates. For example, in the of key vertebrate models—frog, chicken, and time it takes a mouse embryo to reach its second mouse. division, a chicken egg has generated a 60,000- If, instead, we align development at the phy- cell blastodisc and is preparing to gastrulate. At lotypic stage (O’Farrell et al. 2004), one is struck 3 d, the still small mouse embryo has formed by the close relationships among the embryos of a blastocyst with 8 to 10 inner cell mass cells, these model vertebrate organisms, as was von which will contribute to the embryo proper. Baer two centuries ago. Moving from this point But these cells are far from ready to begin devel-

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 15 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

Ancestral program

*

Gastrulation

Phylotypic embryo

Mammalian program

Figure 1. Insertion of a new module of development distinguishes early mammalian embryogenesis from that of its predecessors. Most animals deposit very large eggs in the external environment where they develop into a feeding organism without nutrition or growth. Cleavage cell cycles divide the egg into blastomeres that imme- diately initiate pattern formation and differentiation to produce an embryo, called phylotypic, because all of the species of a phylum look similar at this stage. Even the embryos of other phyla are similar in shape and size at this stage. Mammalian embryogenesis, as exemplified by the mouse, begins with a much smaller egg (magnified somewhat here) and a distinctly different early program of development. The cells resulting from the initial divisions do not gastrulate. First, extra embryonic tissues are produced as the embryo grows enormously and rapidly. Only then does it gastrulate and establish strong parallels to the ancestral program of embryogenesis. The mammalian module of development (boxed) is inserted early during development () so that the earliest divisions are dramatically changed and many of the attributes of the early events of the ancestral program are deferred along with gastrulation.

opment at this point. Over the course of the next malian egg is part of the inserted program; day and a half, the blastocyst hatches out of its hence, it is not surprising that its features protective shell (zona pellucida) and implants. bear little resemblance to the features of early On hatching, it begins to grow rapidly. During divisions in externally deposited eggs. This ar- the next 3 d, the embryo grows in volume 500- rangement suggests that the rapid cleavage fold (Snow 1981). There is also an impressive cycles of the ancestral program have been dis- increase in cell number arriving at 15,000 cells placed and are part of the growth program, a at the time of gastrulation. Although this catch- position that seems incongruous for “cleavage up phase of growth makes a mammalian gastru- cycles,” but we will see that, although there are la that is comparable in size and cell number to important modifications to the early division the gastrulae formed by externally deposited program, this alignment fits numerous obser- eggs, it also changes the foundation for gastru- vations. lation. It diminishes maternal input into em- bryogenesis and separates the cleavage cycles Realignment of Early Development from the onset of patterning. To understand the relationship of the mam- According to the proposed displacement of ear- malian program to the ancestral program, it is ly developmental events to a later stage in mam- important to understand at what point the mals, we should be able to identify parallels be- mammalian module has been inserted. I suggest tween the proliferative division that amplify cell that this module was inserted near the very be- number in the mammalian embryo and the ginning (see asterisk on the ancestral program in cleavage cell cycle. One parallel is timing with Fig. 1). Accordingly, the “cleavage” of the mam- respect to conserved features of embryogene-

16 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

sis—both types of divisions immediately pre- the cycle is regulated unlike other cultured cede gastrulation; another is speed. mammalian cells. During G1, the canonical reg- The cell cycle just before gastrulation of the ulators of progress to S phase, Rb, and cyclin E, mammalian embryo are the fastest known in are inappropriately phosphorylated and abun- mammals. A 5.1-h cell-cycle time would be dant, respectively (White et al. 2005). Further- needed to explain the increasing cell number if more, oscillations of numerous key cell-cycle growth were uniform. However, proliferation is regulators are greatly dampened, such that there not uniform, and, based on mitotic index, Snow is peculiarly high cyclin:Cdk activity in inter- estimated that cells just anterior to the streak, the phase (Savatier et al. 1996; Fujii-Yamamoto morphological marker of gastrulation, double et al. 2005; Yang et al. 2011; van der Laan et al. every 2–3 h (Snow 1977). Rat embryos show a 2013). Although these features of the ES cell similarly fast cell-cycle rate of 3–3.5 h near cycle have puzzled investigators expecting an the streak (Mac Auley et al. 1993). Thus, these invariant mode of cell-cycle regulation, they cycles, like the cleavage cycles of externally de- are familiar to those of us exploring the changes posited eggs, are exceptionally fast. In addition in cell-cycle regulation during early develop- to being fast, other similarities that I have previ- ment in other organisms. Notably, work in Dro- ously reviewed suggest parallels between these sophila has shown that the early rapid cycles are pregastrulation cell cycles in mammals and the associated with very high interphase cyclin:Cdk early cleavage cycles (O’Farrell et al. 2004). Here, activity, and near absence of oscillation in the I present the relationship between mammalian abundance of key cell-cycle regulators (Edgar embryonic stem (ES) cells and the inserted et al. 1994). Furthermore, there is not just one growth phase, as well as describing apparent divergent mode of regulating the early cell cycle parallels of ES cell cycles to early embryonic cy- in Drosophila. As development progresses, the cles in externally deposited eggs. mode of regulating cell-cycle changes in several ES cells were derived from the ICM of mouse steps as S phase slows, a G2 is added and then a blastocysts (Martin 1981). As Martin suggested G1 is introduced. Rb and cyclin E come to play in her report of the successful isolation of ES important regulatory roles only late in this pro- cells, perhaps the cells “represent a selected pop- cess, consistent with their behavior in ES cells. I ulation of completely normal embryonic cells propose that cell-cycle regulation in ES cells, as that are programmed to divide until they receive well as the embryonic cells from which they are the appropriate signals for differentiation.” In- derived, is modified, and these modifications deed, the ICM cells divide and grow during the reflect their derivation from the earlyembryonic rapid-growth phase of the implanted mamma- cycles of externally deposited eggs. It will be lian embryo, and these cells are still growing at interesting to learn the extent of the parallels. the egg cylinder stage, which also successfully In summary, it appears that the mammalian yields pluripotent stem cells (Evans and Kauf- module of development was inserted within the man 1981). Thus, growing ES cells provide a early rapid cell divisions with modifications to model for the growth phase of the pregastrula- these divisions. This insertion displaced many tion mammalian embryo. developmental events to a later stage. ES cells also have a peculiar cell cycle that reverts to more a typical mode of regulation Pluripotency and the Transfer of when these cells are induced to differentiate (Sa- Responsibility to the Zygotic Program vatier et al. 1996). Before differentiation, ES cells divide quickly for a cultured cell, although not Perhaps the chief modification of the early em- nearly as fast as the estimates for the embryonic bryonic programs is a switch from maternal pro- cells. The ES cells also have remarkably short G1 graming to zygotic programming of the rapid and G2 phases, reminiscent of the absence of G1 divisions in mammals. The maternal program- and G2 in the cleavage cycles of externally depos- ming of cleavage divisions used by externally ited eggs (Coronado et al. 2013). Furthermore, deposited eggs is not practical in mammals be-

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 17 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

cause the rapid divisions that continue after ex- several nonmammalian vertebrates, and they are tensive growth would have diluted maternal expressed in early pregastrulation embryos. gene products. Furthermore, mammals have lit- As a result of maternal supply and very early tleneed torelyonmaternalprogramming.Given zygotic expression, the zebrafish homologs of that their eggs are small and that the protected Oct4, Sox2, and Nanog are present in the early mammalian egg developsataleisurely pace,even embryo. They function jointly to promote early a single nucleus can modify transcript levels fast transcription. In their absence, transcription of enough to regulate events. Indeed, mammalian many genes fails and the embryo does not gas- embryos become dependent on zygotic gene ex- trulate (Lee et al. 2013). These factors appear to pression within the first few divisions with ki- act widely to direct transcription of genes nec- netics that differ slightly between species. This essary to stimulate the developmental potential early zygotic program faces requirements absent of the blastomeres of the zebrafish embryo (Lee in externally deposited eggs. The zygotic pro- et al. 2013; Leichsenring et al. 2013). These find- gram of mammalian embryos sustains a grow- ings suggest analogies between the pluripotency ing population of pluripotent cells and then of mammalian embryonic cells and transcrip- provides these cells with a zygotic mimic of the tion factors directing the onset of the initial ancestral maternally run cleavage program. zygotic program of gene expression in fish. The addition of the zygotically supported Despite relatedness, Oct4, Sox2, and Nanog embryonic growth program in mammalian de- have somewhat different jobs in fish and mam- velopment created a new need for a mechanism mals. In fish, these transcription factors act as a to sustain pluripotency. In externally deposited kick-starter to activate the first set of genes that eggs, the early rapid divisions that produce the are transcribed as the embryo transitions to zy- blastomeres appear to be part of the program gotic expression. But the initial wave of tran- that refreshes chromatin structures to create a scription is transient as the embryo immediate- naı¨ve state for the beginning of development. ly advances to gastrulation and differentiation Because these newly produced blastomeres im- and new tiers of zygotic gene expression. In mediately begin morphogenesis and differenti- mammals, the role of pluripotency factors is ation at the time that they initiate zygotic gene less transient. These transcription factors are re- expression, they have no need to maintain their quired to sustain developmental potency during pluripotency. But the mammalian embryo has growth and proliferation until the time of the to zygotically create pluripotency and maintain deferred activation of gastrulation and differen- it during a growth phase that amplifies cell num- tiation. Presumably, vertebrates coopted a pre- ber .500-fold before initiating embryogenesis. existing transcription program that drove early Tremendous advances in the study of stem gene expression in vertebrate ancestors, and re- cells have identified mammalian factors that can configured the wiring of the program so that the reprogram differentiated cells to pluripotency expression of these early activators, their targets, (Takahashi and Yamanaka 2006). These pluri- and the early properties of the cells persist dur- potency factors are expressed in normal embry- ing the growth phase. onic cells, and sustain pluripotency as the ICM In addition to simple persistence, the pluri- cells of the blastocyst grow and proliferate before potency factors may have another job. In con- gastrulation (Marandel et al. 2012; Le Bin et al. trast to fish eggs, which are preloaded with 2014; Sun et al. 2014). If the pluripotent cells of maternal products to run the rapid embryonic the mammalian embryo are analogous to the cycles, the mammalian embryo has to support cleavage stage blastomeres of externally depos- its dramatic growth by zygotic gene expression. ited eggs, perhaps homologs of the pluripotency Perhaps early embryonic transcription factors factors will be expressed in the blastomeres in took on this growth-promoting role and now the predecessors of mammals. Oct4, Sox2, and stimulate expression of cell-cycle and growth Nanog, three factors involved in maintenance of genes to drive the rampant pregastrulation pluripotency in mammals, do have homologs in growth in mammals.

18 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

The presence of the “pluripotency factors” isms called endoparasites that deposit their eggs in cleavage stage zebrafish embryos might sug- within the bodies of other organisms. This life- gest conservation of the program that erases style has evolved independently numerous epigenetic marks. However, relatively rapid di- times (Grbic´ et al. 1998; Grbic´ 2000; Wiegmann vergence of these transcription factors outside et al. 2011), and a diversity of changes in early of mammals shows that pluripotency in other development are found among the species with species can be accomplished independent of this lifestyle (Grbic´ et al. 1998; Grbic´ 2000; these factors. I suggest that the extremely fast Wiegmann et al. 2011). Although perhaps not maternally run divisions in externally deposited an appealing analogy, the life histories of endo- eggs makes a major contribution to erasing epi- parasitic wasps provides a number of indepen- genetic marks, and that the pluripotency factors dent examples of branches of evolution in which took over this role in conjunction with the eggs are released of their responsibility for pro- abandonment of the maternally driven division viding for the development to a feeding adult program in mammals. stage. Additionally, the changes in development associated with these events can be analyzed in some detail because close relatives of the endo- Releasing the Growth Constraint Unleashes parasitic wasps are known and these adhere Evolutionary Change closely to a canonical program of early develop- I have argued that the simple requirement that ment described in detail in Drosophila. an egg be able to produce a feeding organism The endoparasitic wasps lay their eggs inside gave rise to constraints on the growth program the larvae of bigger insects where they develop in of animals. In particular, it forced the large-egg the nutrient-rich hemolymph (Grbic´ et al. 1998; paradigm, and the requirement for a large egg Grbic´ 2000; Wiegmann et al. 2011). The eggs of generated new constraints that underlie the cu- such wasps are small, in the same size range as a rious features of growth and proliferation in mammalian egg, and lack yolk. The early devel- phases 1 and 2. A viviparous lifestyle opens up opment of these small eggs differs dramatically new possibilities. Development of the egg in a from the canonical insect program. Notably, a protected uterine environment in which nu- new developmental module is inserted at the trients can be provided frees evolution of these beginning of development. The early mitotic growth constraints. Once freed of constraints, cycles are greatly changed, and patterning events change is possible and to some extent inevitable. and gastrulation are delayed. During this period, Several of the major changes seen in early the embryo grows dramatically reaching the mammalian development might be explained same size as the externally developing embryos by the release of the constraint on growth. that beganwithvastly biggereggs. The morphol- Clearly, growth of the embryo following hatch- ogy of the gastrulating embryo differs from its ing from the zona pellucida depends on the close relatives; nonetheless, the endoparasitic release of the growth constraint. It also seems wasps express classical insect-patterning genes likely that the small size of the mammalian egg in canonical patterns during this stage and gen- and absence of yolk accumulation were allow- erate a phylotypic embryo resembling that of able changes because of the release of the growth its relatives. Although there is great diversity constraint. But because many coupled changes in the details, the overall features of the changes occurred during the evolution of mammals, it is are similar in various lineages and are not unlike hard to argue that release of the growth con- the reorganization of early embryogenesis in straint was the driving event. Perhaps other ex- mammals. Notably, it is clear that development amples of a release of the growth constraint can is appropriately aligned at the phylotypic stage, give us additional perspective on this. and that a direct comparison of the cleavage It turns out that the constraint on growth of cycles between even closely related species the embryo has been released numerous times would give the impression that there was no re- in evolution, perhaps most frequently in organ- lationship.

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 19 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

Many endoparasitic wasps lay only one egg the egg of the wasp, the egg of the fluke manages in an individual host larva, but, in a number of to produce many embryos—polyembryony. species, numerous progeny eventually emerge However, instead of simple serial twining of form this parasitized larvae. This is the result the embryo, a complex zygotic program of of a remarkable adaptation in which the embryo development in the molluskan host has been twins repeatedly in a process called polyembry- added to zygotically generate many egg-like cells ony (Grbic´ et al. 1998; Grbic´ 2000; Wiegmann that then initiate embryogenesis. This program et al. 2011). This process, which is particularly suggests that, once the growth constraint is re- dramatic and well studied in the wasp Copido- leased, the earliest stage of development is evo- soma floridanum, takes advantage of the growth lutionarily plastic. In contrast to the dramatic stage that was inserted into the early embry- changes in early zygotic development, events onic programs of the endoparasitic wasps. The close to the phylotypic stage persist although pregastrulation embryo of this species forms a in a context divorced from the usually close as- hollow ball of growing and proliferating cells. sociation with fertilization and initial division The shell of cells subdivides to produce addi- of the egg. tional balls as it grows. This process can produce It is interesting to consider pluripotency in in the range of 2000 embryos before patterning, the context of polyembryony. This lifestyle re- gastrulation, and differentiation begin. This quires that embryonic cells grow and proliferate program, which is of obvious advantage in but still retain pluripotency. The transcription producing more wasps, is also an indication of factors responsible for zygotic maintenance of the flexibility of the early developmental pro- pluripotency in mammals are not conserved in gram once the growth constraint is removed. insects, which appear to use different transcrip- Although the relatively close relationship of tion factors (e.g., zelda in Drosophila)topro- endoparasitic wasps with other wasps provides mote early embryonic transcription. Perhaps a particularly clear illustration of the conse- the endoparasitic wasps will have independently quences of the release of the growth constraint, evolved a means of supporting pluripotency other parasites provide additional examples of during growth and proliferation. For example, modification of the ancestral early embryonic persistent expression of a Zelda-like transcrip- program. Flukes, or trematodes, comprise a sub- tion factor might provide a wasp embryo with stantial class of parasitic flatworms with about the equivalent of a pluripotency factor. We four times as many species as there are species might expect that ES cell lines could be estab- of mammals. Usually, the primary or defini- lished from these organisms (see Jourdane and tive host is a vertebrate. The parasites, often her- Theron 1980). maphroditic, produce small encysted eggs that Overall, the rapid evolution of the early de- are released from the infected primary host velopmental programs in endoparasitic wasps (Butcher et al. 2002). Typically, these eggs, if supports the interpretation that the growth con- ingested by a mollusk, will develop into a prim- straint played a major role in restraining such itive larva called a miracidium that takes in divergence. Despite occasional release of the nourishment from the host mollusk and grows constraint, we expect that many aspects of the and develops germinative tissue that generates regulatory programs that control growth and egg-like cells that enlarge and go through stages proliferation were molded by an unchanging recognizable as cleavage, gastrulation, and es- constraint that influenced much of animal evo- tablishment of a phylotypic body plan (Pod- lution. vyaznaya and Galaktionov 2014; Ska´la et al. 2014). Notably, growth occurs throughout SUMMARY this embryogenesis. The miracidium nourishes many embryos in a uterus-like sac. On release, Consideration of oogenesis and early develop- the progeny, now called cercariae, can infect ment across awide swath of the animal kingdom the vertebrate host to begin a new cycle. Like suggests that the issue of growth has been an

20 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

enduring constraint influencing much of evolu- REFERENCES tion. It is suggested that this constraint emerges Bedford L. 1966. The electron microscopy and cytochemis- from a simple requirement that an externally try of oogenesis and the cytochemistry of embryonic de- deposited egg be able to produce a feeding an- velopment of the prosobranch gastropod Bembicium na- imal to have a workable life cycle. This demand num L. J Embryol Exp Morphol 15: 15–37. was met by production of large eggs. This then Blobel G, Potter VR. 1967. Studies on free and membrane- bound ribosomes in rat liver. I: Distribution as related to produced the widespread pattern of cell growth total cellular RNA. J Mol Biol 26: 279–292. in phase 1 (oocyte production) and cell prolif- Bratic I, Hench J, Henriksson J, Antebi A, Bu¨rglin TR, Tri- eration in phase 2 (early embryogenesis) of an funovic A. 2009. Mitochondrial DNA level, but not active replicase, is essential for Caenorhabditis elegans develop- overall four-phase animal program of growth ment. Nucleic Acids Res 37: 1817–1828. and proliferation. This solution stressed the ca- Brown DD, Dawid IB. 1968. Specific gene amplification in pacities of ordinary growth and division. The oocytes. Oocyte nuclei contain extrachromosomal repli- growth of an oocyte and its division as an egg cas of the genes for ribosomal RNA. Science 160: 272– 280. shows numerous accommodations to these Butcher AR, Brealey JK, Grove DI, Dymock RB. 2002. Bra- stresses. These accommodations are evident in chylaima cribbi (Digenea: Brachylaimidae): Scanning representatives of all animal phyla, arguing that electron microscopical observations of the life-cycle stag- the biology of these stages has been molded es. J Helminthol 76: 207–215. Can˜on S, Fernandez-Tresguerres B, Manzanares M. 2011. by the pressures of this enduring constraint. Pluripotency and lineages in the mammalian blastocyst: Furthermore, embryonic development evolved An evolutionary view. Cell Cycle 10: 1731–1738. with these imposed constraints, so the pro- Chase JW,Dawid IB. 1972. Biogenesis of mitochondria dur- grams of development are interwoven with ing Xenopus laevis development. Dev Biol 27: 504–518. the programs for growth and proliferation. Fi- Coronado D, Godet M, Bourillot PY, Tapponnier Y, Bernat A, Petit M, Afanassieff M, Markossian S, Malashicheva A, nally, I argue that mammalian development Iacone R, et al. 2013. A short G1 phase is an intrinsic ought to be viewed as a diverged branch of determinant of naı¨ve embryonic stem cell pluripotency. the mainline of animal evolution, which was Stem Cell Res 10: 118–131. dramatically influenced by the release of the Cox RT, Spradling AC. 2003. A Balbiani body and the fu- some mediate mitochondrial inheritance during Droso- growth constraint that was originally forma- phila oogenesis. Development 130: 1579–1590. tive. Introduction of an early growth phase in Del Pino EM, Humphries AA. 1978. Multiple nuclei during mammalian embryogenesis required a novel early oogenesis in Flectonotus pygmaeus and other mar- feature, persistent pluripotency of the prolifer- supial frogs. Biol Bull 154: 198–212. Dobzhansky T.1973. Nothing in biology makes sense except ating cells, which go on to ultimately support in the light of evolution. JHered68: 3–10. the deferred embryogenesis. Parasitic organ- Domazet-Losˇo T, Tautz D. 2010. A phylogenetically based isms, which also evade the growth constraint, transcriptome age index mirrors ontogenetic divergence show diverged early programs including small patterns. Nature 468: 815–818. Eckelbarger KJ, Larson R. 1992. Ultrastructure of the ovary eggs, early embryonic growth, and acquisition and oogenesis in the jellyfish Linuche unguiculata and of a proliferating pluripotent population of Stomolophus meleagris, with a review of ovarian structure cells. These examples of the release of the growth in the Scyphozoa. Marine Biol 114: 633–643. constraint are an indication of the powerful in- Eckelbarger KJ, YoungCM. 1997. Ultrastructure of the ovary and oogenesis in the methane-seep mollusc Bathynerita fluence that this constraint has exerted on ani- naticoidea (Gastropoda: Neritidae) from the Louisiana mal evolution. slope. Invertebr Biol 116: 299–312. Edgar BA, O’Farrell PH. 1990. The three postblastoderm cell cycles of Drosophila embryogenesis are regulated in G2 by ACKNOWLEDGMENTS string. Cell 62: 469–480. Edgar BA, Schubiger G. 1986. Parameters controlling tran- I thank current and past members of the labo- scriptional activation during early Drosophila develop- ratory, especially Jeffrey Farrell, Jennifer Page, ment. Cell 44: 871–877. and Isaac Strong, as well as Danielle O’Farrell Edgar BA, Kiehle CP, Schubiger G. 1986. Cell cycle control by the nucleo-cytoplasmic ratio in early Drosophila de- for help in preparing this article. Our efforts velopment. Cell 44: 365–372. are supported by Grants R37GM037193 and Edgar BA, Sprenger F, Duronio RJ, Leopold P,O’Farrell PH. R01ES020725 to P.H.O. 1994. Distinct molecular mechanism regulate cell cycle

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 21 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

timing at successive stages of Drosophila embryogenesis. Gurdon JB, Elsdale TR, Fischberg M. 1958. Sexually mature Genes Dev 8: 440–452. individuals of Xenopus laevis from the transplantation of Evans MJ, Kaufman MH. 1981. Establishment in culture of single somatic nuclei. Nature 182: 64–65. pluripotential cells from mouse embryos. Nature 292: Harvey EB. 1935. Cleavage without nuclei. Science 82: 277. 154–156. Hayashi S, YamaguchiM. 1999. Kinase-independent activity Evans T,Rosenthal ET,YoungblomJ, Distel D, Hunt T.1983. of Cdc2/cyclin A prevents the S phase in the Drosophila Cyclin: A protein specified by maternal mRNA in sea cell cycle. Genes Cells 4: 111–122. urchin eggs that is destroyed at each cleavage division. Howe JA, Howell M, Hunt T, Newport JW. 1995. Identifica- Cell 33: 389–396. tion of a developmental timer regulating the stability of Farrell JA, O’Farrell PH. 2013. Mechanism and regulation of embryonic cyclin A and a new somatic A-type cyclin at Cdc25/twine protein destruction in embryonic cell-cycle gastrulation. Genes Dev 9: 1164–1176. remodeling. Curr Biol 23: 118–126. Jourdane J, Theron A. 1980. Schistosoma mansoni: Cloning Farrell JA, O’Farrell PH. 2014. From egg to gastrula: How the by microsurgical transplantation of sporocysts. Exp Para- cell cycle is remodeled during the Drosophila mid-blas- sitol 50: 349–357. tula transition. Annu Rev Genet 48: 269–294. Kalinka AT, Varga KM, Gerrard DT, Preibisch S, Corcoran Follette PJ, O’Farrell PH. 1997. Connecting cell behavior to DL, Jarrells J, Ohler U, Bergman CM, Tomancak P.2010. patterning: Lessons from the cell cycle. Cell 88: 309–314. Gene expression divergence recapitulates the develop- Follette PJ, Duronio RJ, O’Farrell PH. 1998. Fluctuations in mental hourglass model. Nature 468: 811–814. cyclin E levels are required for multiple rounds of endo- Kloc M, Jedrzejowska I, Tworzydlo W, Bilinski SM. 2014. cycle S phase in Drosophila. Curr Biol 8: 235–238. Balbiani body, nuage and sponge bodies—The germ Fujii-Yamamoto H, Kim JM, Arai KI, Masai H. 2005. Cell plasm pathway players. Arthropod Struct Dev 43: 341– cycle and developmental regulations of replication fac- 348. tors in mouse embryonic stem cells. J Biol Chem 280: Landstro¨m U, Løvtrup-Rein H, Løvtrup S. 1975. Control of 12976–12987. cell division and cell differentiation by deoxynucleotides in the early embryo of Xenopus laevis. Cell Differ 4: 313– Gall JG. 1968. Differential synthesis of the genes for ribo- 325. somal RNA during amphibian oo¨genesis. Proc Natl Acad Sci 60: 553–560. Laskey RA, Honda BM, Mills AD, Finch JT. 1978. Nucleo- somes are assembled by an acidic protein which binds Gall JG. 2012. Are lampbrush chromosomes unique to mei- histones and transfers them to DNA. Nature 275: 416– otic cells? Chromosome Res 20: 905–909. 420. Gall JG, Wu Z, Murphy C, Gao H. 2004. Structure in the Le Bin GC, Mun˜oz-Descalzo S, Kurowski A, Leitch H, Lou amphibian germinal vesicle. Exp Cell Res 296: 28–34. X, Mansfield W, Etienne-Dumeau C, Grabole N, Mulas Ganot P,Bouquet JM, Kallesøe T,Thompson EM. 2007. The C, Niwa H, et al. 2014. Oct4 is required for lineage prim- Oikopleura coenocyst, a unique chordate germ cell per- ing in the developing inner cell mass of the mouse blas- mitting rapid, extensive modulation of oocyte produc- tocyst. Development 141: 1001–1010. tion. Dev Biol 302: 591–600. Lee MT, Bonneau AR, Takacs CM, Bazzini AA, DiVito KR, Ganot P,Moosmann-Schulmeister A, Thompson EM. 2008. Fleming ES, Giraldez AJ. 2013. Nanog, Pou5f1 and SoxB1 Oocyte selection is concurrent with meiosis resumption activate zygotic gene expression during the maternal-to- in the coenocystic oogenesis of Oikopleura. Dev Biol 324: zygotic transition. Nature 503: 360–364. 266–276. Leichsenring M, Maes J, Mo¨ssner R, Driever W, Onicht- Gardner EJ, Nizami ZF, Talbot CC, Gall JG. 2012. Stable chouk D. 2013. Pou5f1 transcription factor controls zy- intronic sequence RNA (sisRNA), a new class of noncod- gotic gene activation in vertebrates. Science 341: 1005– ing RNA from the oocyte nucleus of Xenopus tropicalis. 1009. Genes Dev 26: 2550–2559. Li XY, Harrison MM, Villalta JE, Kaplan T, Eisen MB. 2014. Gersh ES. 1968. Mutants at the bobbed locus in Drosophila Establishment of regions of genomic activity during the melanogaster: Relation to ribosomal RNA synthesis? Sci- Drosophila maternal to zygotic transition. eLife doi: 10. ence 162: 1139. 7554/eLife.03737. Graham A, Richardson J. 2012. Developmental and evolu- Lin CJ, Koh FM, Wong P, Conti M, Ramalho-Santos M. tionary origins of the pharyngeal apparatus. Evodevo 3: 2014. Hira-mediated H3.3 incorporation is required for 24. DNA replication and ribosomal RNA transcription in the Grbic´ M. 2000. “Alien” wasps and evolution of develop- mouse zygote. Dev Cell 30: 268–279. ment. Bioessays 22: 920–932. Loppin B, Bonnefoy E, Anselme C, Laurenc¸on A, Karr TL, Grbic´ M, Nagy LM, Strand MR. 1998. Development of poly- Couble P. 2005. The histone H3.3 chaperone HIRA is embryonic insects: A major departure from typical insect essential for chromatin assembly in the male pronucleus. embryogenesis. Dev Genes Evol 208: 69–81. Nature 437: 1386–1390. Gruzova MN, Batalova FM. 1979. Nuclear structures in the Mac Auley A, Werb Z, Mirkes PE. 1993. Characterization of telotrophic ovarioles of nocturnal ground beetles (Tene- the unusually rapid cell cycles during rat gastrulation. brionidae, Polyphaga). II: The oocyte nucleus of Blaps Development 117: 873–883. lethifera and Gnaptor spinimanus. Light optical data. On- Macgregor HC, Del Pino EM. 1982. Ribosomal gene ampli- togenez 10: 323–331. fication in multinucleate oocytes of the egg brooding

22 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell

hylid frog Flectonotus pygmaeus. Chromosoma 85: 475– Prud’homme B, de Rosa R, Arendt D, Julien JF, Pajaziti R, 488. Dorresteijn AWC, Adoutte A, Wittbrodt J, Balavoine Marandel L, Labbe C, Bobe J, Le Bail PY. 2012. nanog 50- G. 2003. Arthropod-like expression patterns of en- upstream sequence, DNA methylation, and expression in grailed and wingless in the annelid Platynereis dumerilii gametes and early embryo reveal striking differences be- suggest a role in segment formation. Curr Biol 13: 1876– tween teleosts and mammals. Gene 492: 130–137. 1881. Marlow FL, Mullins MC. 2008. Bucky ball functions in Bal- Raff RA. 1996. The shape of life. University of Chicago biani body assembly and animal–vegetal polarity in the Press, Chicago. oocyte and follicle cell layer in zebrafish. Dev Biol 321: Rahn H, Paganelli CV,Ar A. 1974. The avian egg: Air-cell gas 40–50. tension, metabolism and incubation time. Respir Physiol Martin GR. 1981. Isolation of a pluripotent cell line from 22: 297–309. early mouse embryos cultured in medium conditioned Sa´nchez-Sa´nchez AV, Camp E, Mullor JL. 2011. Fishing by teratocarcinoma stem cells. Proc Natl Acad Sci 78: pluripotency mechanisms in vivo. Int J Biol Sci 7: 410– 7634–7638. 417. Newport J, Dasso M. 1989. On the coupling between DNA Savatier P, Lapillonne H, van Grunsven LA, Rudkin BB, replication and mitosis. J Cell Sci 12: 149–160. Samarut J. 1996. Withdrawal of differentiation inhibitory Newport J, Kirschner M. 1982a. A major developmental activity/leukemia inhibitory factor up-regulates D-type transition in early Xenopus embryos. I: Characterization cyclins and cyclin-dependent kinase inhibitors in mouse and timing of cellular changes at the midblastula stage. embryonic stem cells. Oncogene 12: 309–322. Cell 30: 675–686. Shen MM. 2007. Nodal signaling: Developmental roles and Newport J, Kirschner M. 1982b. A major developmental regulation. Development 134: 1023–1034. transition in early Xenopus embryos. II: Control of the Shermoen AW, O’Farrell PH. 1991. Progression of the cell onset of transcription. Cell 30: 687–696. cycle through mitosis leads to abortion of nascent tran- O’Farrell PH. 2004. How metazoans reach their full size: The scripts. Cell 67: 303–310. natural history of bigness. In Cell growth: Control of cell Shermoen AW, McCleland ML, O’Farrell PH. 2010. Devel- size (ed. Hall M, et al.). Cold Spring Harbor Laboratory opmental control of late replication and S phase length. Press, Cold Spring Harbor, NY. Curr Biol 20: 2067–2077. O’Farrell PH. 2011. Quiescence: Early evolutionary origins Ska´la V,Bulantova´ J, Walker AJ, Hora´k P.2014. Insights into and universality do not imply uniformity. Phil Trans R the development of Notocotylus attenuatus (Digenea: No- Soc B 366: 3498–3507. tocotylidae) in Lymnaea stagnalis: From mother sporo- O’Farrell PH, Stumpff J, Su TT. 2004. Embryonic cleavage cyst to cercariae. Parasitol Int 63: 94–99. cycles: How is a mouse like a fly? Curr Biol 14: R35–R45. Smiley S. 1990. A review of echinoderm oogenesis. J Electron Orsi GA, Algazeery A, Meyer RE, Capri M, Sapey-Triomphe Microsc Tech 16: 93–114. LM, Horard B, Gruffat H, Couble P,Aı¨t-Ahmed O, Lop- Snow MHL. 1977. Gastrulation in the mouse: Growth and pin B. 2013. Drosophila Yemanuclein and HIRA cooper- regionalization of the epiblast. J Embryol Exp Morphol 42: ate for de novo assembly of H3.3-containing nucleo- 293–303. somes in the male pronucleus. PLoS Genet 9: e1003285. Snow MH. 1981. Growth and its control in early mamma- Patel NH, Martin-Blanco E, Coleman KG, Poole SJ, Ellis lian development. Br Med Bull 37: 221–226. MC, Kornberg TB, Goodman CS. 1989. Expression of Sun LT,YamaguchiS, Hirano K, Ichisaka T,Kuroda T,TadaT. engrailed proteins in arthropods, annelids, and chor- 2014. Nanog co-regulated by Nodal/Smad2 and Oct4 is dates. Cell 58: 955–968. required for pluripotency in developing mouse epiblast. Pepling ME, de Cuevas M, Spradling AC. 1999. Germline Dev Biol 392: 182–192. cysts: A conserved phase of germ cell development? Takahashi K, Yamanaka S. 2006. Induction of pluripotent Trends Cell Biol 9: 257–262. stem cells from mouse embryonic and adult fibroblast Philpott A, Leno GH, Laskey RA. 1991. Sperm decondensa- cultures by defined factors. Cell 126: 663–676. tion in Xenopus egg cytoplasm is mediated by nucleoplas- Tsang WY, Lemire BD. 2002. Mitochondrial genome con- min. Cell 65: 569–578. tent is regulated during nematode development. Biochem Philpott A, Krude T, Laskey RA. 2000. Nuclear chaperones. Biophys Res Commun 291: 8–16. Semin Cell Dev Biol 11: 7–14. Tworzydlo W,Kloc M, Bilinski SM. 2009. The Balbiani body Podvyaznaya IM, Galaktionov KV. 2014. Trematode repro- in the female germline cells of an earwig, Opisthocosmia duction in the molluscan host: An ultrastructural studyof silvestris. Zool Sci 26: 754–757. the germinal mass in the rediae of Himasthla elongata van der Laan S, Tsanov N, Crozet C, Maiorano D. 2013. High (Mehlis, 1831) (Digenea: Echinostomatidae). Parasitol Dub3 expression in mouse ESCs couples the G1/S check- Res 113: 1215–1224. point to pluripotency. Mol Cell 52: 366–379. Pritchard DK, Schubiger G. 1996. Activation of transcrip- Webb AC, Smith LD. 1977. Accumulation of mitochondrial tion in Drosophila embryos is a gradual process mediated DNA during oogenesis in Xenopus laevis. Dev Biol 56: by the nucleocytoplasmic ratio. Genes Dev 10: 1131– 219–225. 1142. White J, Stead E, Faast R, Conn S, Cartwright P, Dalton S. Prud’homme B, Gompel N. 2010. Evolutionary biology: 2005. Developmental activation of the Rb-E2F pathway Genomic hourglass. Nature 468: 768–769. and establishment of cell cycle-regulated cyclin-depen-

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 23 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

P.H. O’Farrell

dent kinase activity during embryonic stem cell differen- Woodland HR, Pestell RQ. 1972. Determination of the nu- tiation. Mol Biol Cell 16: 2018–2027. cleoside triphosphate contents of eggs and oocytes of Wiegmann BM, Trautwein MD, Winkler IS, Barr NB, Kim Xenopus laevis. Biochem J 127: 597–605. JW, Lambkin C, Bertone MA, Cassel BK, Bayless Yang VS, Carter SA, Hyland SJ, Tachibana-Konwalski K, KM, Heimberg AM, et al. 2011. Episodic radiations Laskey RA, Gonzalez MA. 2011. Geminin escapes degra- in the fly tree of life. Proc Natl Acad Sci 108: dation in G1 of mouse pluripotent cells and mediates the 5690– 5695. expression of Oct4, Sox2, and Nanog. Curr Biol 21: 692– Wieschaus E. 1996. Embryonic transcription and the con- 699. trol of developmental pathways. Genetics 142: 5–10. Yuan K, Shermoen AW, O’Farrell PH. 2014. Illuminating Wolf SF, Schlessinger D. 1977. Nuclear metabolism of DNA replication during Drosophila development using ribosomal RNA in growing, methionine-limited, and TALE-lights. Curr Biol 24: R144–R145. ethionine-treated HeLa cells. Biochemistry 16: 2783– Zhao ZK, Li W,WangMY,Zhou L, Wang JL, WangYF.2011. 2791. The role of HIRA and maternal histones in sperm nucleus Woodland H. 1982. The translational control phase of early decondensation in the gibel carp and color crucian carp. development. Biosci Rep 2: 471–491. Mol Reprod Dev 78: 139–147.

24 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a019042 Downloaded from http://cshperspectives.cshlp.org/ on September 26, 2021 - Published by Cold Spring Harbor Laboratory Press

Growing an Embryo from a Single Cell: A Hurdle in Animal Life

Patrick H. O'Farrell

Cold Spring Harb Perspect Biol 2015; doi: 10.1101/cshperspect.a019042 originally published online August 7, 2015

Subject Collection Size Control in Biology: From Organelles to Organisms

Cell-Size Control Mechanical Forces and Growth in Animal Tissues Amanda A. Amodeo and Jan M. Skotheim Loïc LeGoff and Thomas Lecuit Indeterminate Growth: Could It Represent the Biological Scaling Problems and Solutions in Ancestral Condition? Amphibians Iswar K. Hariharan, David B. Wake and Marvalee Daniel L. Levy and Rebecca Heald H. Wake The Systemic Control of Growth Intracellular Scaling Mechanisms Laura Boulan, Marco Milán and Pierre Léopold Simone Reber and Nathan W. Goehring Genome Biology and the Evolution of Cell-Size Growing an Embryo from a Single Cell: A Hurdle Diversity in Animal Life Rachel Lockridge Mueller Patrick H. O'Farrell Size Scaling of Microtubule Assemblies in Early Organ-Size Regulation in Mammals Xenopus Embryos Alfredo I. Penzo-Méndez and Ben Z. Stanger Timothy J. Mitchison, Keisuke Ishihara, Phuong Nguyen, et al. The Influence of Genome and Cell Size on Brain Size Control in Plants−−Lessons from Leaves and Morphology in Amphibians Flowers Gerhard Roth and Wolfgang Walkowiak Hjördis Czesnick and Michael Lenhard The Opposing Actions of Target of Rapamycin Nuclear DNA Content Varies with Cell Size across and AMP-Activated Protein Kinase in Cell Growth Human Cell Types Control James F. Gillooly, Andrew Hein and Rachel Sravanth K. Hindupur, Asier González and Michael Damiani N. Hall Small but Mighty: Cell Size and Bacteria Subcellular Size Petra Anne Levin and Esther R. Angert Wallace F. Marshall

For additional articles in this collection, see http://cshperspectives.cshlp.org/cgi/collection/

Copyright © 2015 Cold Spring Harbor Laboratory Press; all rights reserved